squat-detection-application

This is a desktop application that utilizes mediapipe library for pose estimation to detect squats. This application is being deployed in Romania to encourage people to do squats, at the end of a certain successful number of squats (e.g. 5), the vending machine will dispense a coin to the customers.

The application is being developed by pyqt5 library and deployed in an Intel NUC mini pc which maintains serial communication with a coin-dispenser machine. The serial communication protocol can be found here: communication-protocol

Try the app yourself:

• Clone the repository

  git clone https://github.com/arrafi-musabbir/squat-counting-application.git
  

• Go the clone directory and install the requirements

  pip install -r requirements.txt
  

• Run the app

  python app.py
  

or download the app.exe from release-v1

GUI workflow	Squat Detection in Backend

minimum system requirements

• Ubuntu LTS 22.04
• Intel-core-i3-5100u-processor-3m-cache-2.40-ghz
• Included Memory 1x4GB DDR4
• Included Storage 128GB HDD
• Python 3.10.12